Generating a synthetic graph that is similar to a given real-world graph is a critical requirement for privacy preservation and benchmarking purposes. Various generative models attempt to generate static graphs similar to real-world graphs. However, generation of temporal graphs is still an open research area. We present a temporal-motif based approach to generate synthetic temporal graph datasets and results from three real-world use cases.
Revised: April 19, 2019 |
Published: August 20, 2018
Citation
Purohit S., L. Holder, and G. Chin. 2018.Temporal Graph Generation Based on a Distribution of Temporal Motifs. In 14TH INTERNATIONAL WORKSHOP ON MINING AND LEARNING WITH GRAPHS (MLG 2018), August 20, 2018, London, United Kingdom.PNNL-SA-134797.